Approximately 1 out of 10 couples in Kenya is HIV-1 discordant despite ongoing sexual activity that results in HIV-1 exposure for the uninfected member of the couple. Several groups have described cellular immune responses in HIV-1-exposed individuals and have suggested that these contribute to reduced susceptibility to HIV-1 infection. However, none of these studies has been able to determine whether the presence of an HIV-1-specific immune response protects against HIV-1 acquisition or is merely a marker of intermittent viral exposure. In this study, we propose to recruit a large cohort of HIV-1-discordant couples and follow both members of the couple prospectively, using HIV-1 transmission as the endpoint. For Aim 1 we will determine whether positive HIV-1-specific CD8+ interferon-gamma (IFN-gamma) responses in the HIV-1-uninfected partner are associated with protection against HIV-1 acquisition during 24 months of follow-up. Aim 2 focuses on HIV-1- specific CD4+ T helper responses and will ascertain whether these are significantly more common in non- transmitting couples compared to HIV-1 transmitting couples. In Aim 3 we will perform detailed analyses of HIV-1-specific T cell function, phenotype, and durability in a subset of 10 enrolled couples, defining viral specificity by sequencing genital and plasma HIV-1 from infected partners. A total of 600 HIV-1-discordant couples will be enrolled in this Nairobi-based study, with approximately 1/3 of couples recruited from an ongoing discordant couple study and the remaining coming from the community. At enrollment, CD8+ and CD4+ T cell responses against peptides derived from a consensus sequence clade A HIV-1 will be measured in the uninfected partner using ELISpot assays and intracellular cytokine staining. At the same visit, HIV-1-infected partners will have a baseline HIV-1 viral load and genital tract HIV-1 RNA shedding quantified using PCR. Follow-up visits will occur every 3 months for both partners and will include assessment of sexual risk behavior and rapid HIV-1 testing for the uninfected partner. During 2 years of follow-up, we estimate 54-65 (approximately 10%) couples will transmit HIV-1 despite regular HIV-1 prevention counseling. Comparing HIV-1 transmitting with non-transmitting couples will inform future vaccine research about whether such responses are associated with reductions in HIV-1 acquisition and may provide valuable insight into the nature of protective cellular immunity for future vaccine development.